Why Time Span of Discretion Works
Timespan of Discretion is the biggest point of complaint that the old folks in the Glacier Metal related work (Requisite Organization, Stratified Systems Theory, Career Path Appreciation, work levels or worklevels, etc.). If you’re new, you may be scratching your head about this Time Span of Discretion.
Elliott Jaques claimed to have discovered that you could determine the complexity of a work role by measuring how long the incumbent could work on their own, before being checked by the boss; or, if you like, how long will it take for your decision about something to come “due”. Frontline work had lower “timespans of discretion” because the results of their decisions can be seen relatively quickly (days or weeks) while a high level decision may not come due for many years, even decades.
Jaques was no idle chatterer. He was a respected psychoanalyst and organizational consultant, an early member of Tavistock Institute, published in things like Harvard Business Review, and in his spare time coined the phrase “mid-life crisis”.
Still, some of his co-researchers think that time span of discretion is hooey, researchers like Ralph Rowbottom (UK National Health Service), David Billis (Unilever, Tata) and Warren Kinston. “We can’t find timespans longer than a couple of years at these CEO roles!” they told him. So they abandoned it. The old guys at the US Army also said timespan made no sense for the war-fighting army, so they threw it out, too.
Kathryn Cason, Jaques’s widow and co-author, has plainly said, “If you’re not doing time span, you’re not doing Jaques!”
What’s going on? Let’s take a look.
The idea of a timespan for a job is familiar to all of us who have been project-oriented. How long should this person be given something to accomplish? Project managers even develop — of necessity — a sense of how long you can let someone work without checking on them. You develop a list of folks who are “competent”, whom you can rely upon to work effectively within the limits they have.
Managers, however, and especially executives, have a hard time with it.
It’s intuitive, if you think about it. If you are managing someone who is responsible for a ten year project, you should probably have a ten-year timespan on your job. How can I manage someone’s work when my discretion ends before his does?
The problem that Jaques ex-disciples had was that they were following Jaques’s prescriptions for discovering timespan. Never, ever follow the founders way of doing things. They are almost always doing something unconsciously, something that’s part of their internal knowledge. Jaques was.
In the end, you can get timespans from lower roles pretty easily. It is simple to say that the oil exploration techs often work a site for 10-12 years before it can be brought online and pump real oil. The CEO of the energy company must be able to manage strategic uncertainty that far out, at least.
The problem seems to be that these Requisite Organization consultants don’t like working with the little people. They want to work at the same level they are. Which is foolishness, of course. They also rarely have broad enough technical expertise to understand technicians and experts when they talk about their work, as they come out of Organizational Design or Human Resources, or were straight up academics. (There are a few notable exceptions.)
Technical experts working projects that come due in 5+ years often report to managers whose roles are limited to 2 years or less. Often less than a year.
But it should be true. The managerial role should have a longer timespans of discretion than their subordinates, such as managing strategic uncertainty, in order to provide value to their subordinates, to steal Nick Forrest’s phrase.
If you want to know what the timespan of discretion of a role should be, go to the technical experts reporting into it and get the timespans from them. If you have a CIO who thinks that managing five years out is insane but whose enterprise architect has to plan and manage architectures out five years or more, fire the CIO. The EA would be a bad choice but at least could do the thinking required of the job.
Licensed through 123rf.com